Dual Inhibition of Human Leukocyte Elastase and Lipid Peroxidation

11 rue des Moulineaux, 92150 Suresnes, France. Received November 8, 1996X. A series of potent and selective human leukocyte elastase (HLE) inhibitors ...
0 downloads 0 Views 533KB Size
1906

J. Med. Chem. 1997, 40, 1906-1918

Dual Inhibition of Human Leukocyte Elastase and Lipid Peroxidation: In Vitro and in Vivo Activities of Azabicyclo[2.2.2]octane and Perhydroindole Derivatives Bernard Portevin,† Michel Lonchampt,‡ Emmanuel Canet,‡ and Guillaume De Nanteuil*,† Division D of Medicinal Chemistry and Division of Respiratory Pharmacology, Institut de Recherche Servier, 11 rue des Moulineaux, 92150 Suresnes, France Received November 8, 1996X

A series of potent and selective human leukocyte elastase (HLE) inhibitors of the Val-Pro-Val type has been developed. Initially, the central proline residue was replaced by nonnatural amino acids Phi ((2S,3aS,7aS)-perhydroindole-2-carboxylic acid) and Abo ((3S)-2-azabicyclo[2.2.2]octane-3-carboxylic acid), and secondly several groups able to confer antioxidant properties to the molecule were introduced at the lipophilic N-terminal side chain. When compared to reference inhibitors, in vitro HLE inhibitory potency was maintained (10-100 nM) both with compounds containing the antioxidant moiety at the end of the N-terminal side chain and with compounds in which the N-terminal valine of the tripeptidic sequence had been replaced by a -substituted lysine. The lipidic peroxidation inhibitory potency of this series of inhibitors was found to be similar to that of the reference antioxidant compounds (around 1 µM). Moreover, HLE-induced hemorrhage in the hamster lung was effectively prevented (40-60% at 15 µg/ kg) by most of the inhibitors tested when administered intratracheally 3 h before instillation of elastase. Among the most active analogs, compounds 11a,c,g were still active when administered 18 h before elastase. Interestingly, compound 14a was able to prevent HLEmediated lung damage when administered 72 h prior to enzymatic challenge, indicating exceptional stability and retention in the lung. Finally, in a 14-day chronic model of emphysema in the hamster, 14a significantly conserved alveolar spaces, a marker of lung tissue destruction, and was more potent than reference inhibitor ICI 200 880. This indicates that addition of peroxidation inhibitory properties to an HLE inhibitor can provide a powerful in vivo inhibitor of pulmonary tissue destruction. Introduction Human leukocyte elastase (HLE, EC 3.4.21.37) has been considered for over 2 decades as one of the most interesting biological targets in the research field of proteolytic destruction of lung tissue.1 Subsequent to its secretion by polymorphonuclear neutrophils, one of the main actions of HLE is the degradation of structural proteins, including elastin, fibronectin, and collagen. Elastin is a highly cross-linked and flexible protein, mainly present in the macroscopic structure of pulmonary parenchyma. Under physiological conditions, the lung is protected from elastolytic activity by endogenous inhibitors such as R1-protease inhibitor (R1-PI); however, in the course of pathologies such as emphysema, cystic fibrosis, or adult respiratory distress syndrome (ARDS), the balance between the natural inhibitor and HLE is displaced in favor of the enzyme2,3 (Scheme 1). Moreover, oxygen radical species (ORS) have been shown to be implicated in the development of emphysema and ARDS in two ways:4 firstly, through a direct degradative action of cell structure and, secondly, through an indirect oxidation, by inactivation of antiproteases such R1-PI. In fact, ORS is contributing to lung destruction by increasing HLE degradative activity, suggesting that a synergestic action could exist between HLE and ORS. Furthermore, it is known that cigarette smoke is a source of active ORS:5 it was † ‡ X

Division D of Medicinal Chemistry. Division of Respiratory Pharmacology. Abstract published in Advance ACS Abstracts, May 15, 1997.

S0022-2623(96)00772-8 CCC: $14.00

reported to inactivate R1-PI and ceruloplasmine ferroxidase, an important oxygen radical scavenger. When we started our investigations in the early 1990s, the tripeptidic trifluoromethyl ketone ICI 200 880 (or ZD 200 880) (1) was already described as a potent and selective HLE inhibitor.6 Moreover, this compound presented a strong in vivo activity in several animal models of pulmonary lesions.7

Since then, several other pharmaceutical companies involved in research in the respiratory field have patented similar structures to the ICI compound,8,9 all of them being derived from the tripeptidic sequence ValPro-Val. In our ongoing search for new and therapeutically useful low molecular weight inhibitors of serine proteinases, we have recently described the replacement of a proline residue by one of our original nonnatural amino acids in the generic structure of a known inhibitory peptidic sequence.10,11 ICI 200 880 appeared at this time to be a suitable reference molecule in order to validate our structural hypothesis. Moreover, we were interested by introducing in the same molecule an antiHLE activity and an antioxidant activity. With this aim, our chemical approach consisted of coupling modified Val-Pro-Val sequences with different groups able to confer an antioxidant property to the respective molecules. © 1997 American Chemical Society

Dual Inhibition of HLE and Lipid Peroxidation

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12 1907

Scheme 1

I

Scheme 2a

a a: AA ) Phi. b: AA ) Abo. Reagents: (a) isobutyl chloroformate, NMM; (b) Ac2O, DMSO; (c) gaseous HCl, dioxane; (d) BocVal-OH, HOBT, DCC, DMF.

Few compounds appear in the literature with this dual inhibitory capacity. CT-1037 (2) is described as a potent elastase inhibitor (IC50 ) 3 nM) embedded with an oxidative activation feature,12 and P 1517 (3) is given as an elastase inhibitor with an IC50 on HLE around 0.1 mM but with in vivo radical scavenger properties superior to N-acetylcysteine.13

methodology. In contrast to previous papers where oxidation of the alcohol was performed at the very last step of the synthesis of the inhibitors, we were able to prepare the dipeptidic trifluoromethyl ketone iv in satisfactory yields using Moffat’s reagent (Ac2ODMSO). This reagent proved to be superior to other oxidants such as the Swern reagent or Dess-Martin periodinane. Deprotection of the N-terminal amino group of iv to give the dipeptide v was effected by treatment with dioxane, previously saturated with dry hydrogen chloride gas. A second coupling reaction with Boc-valine in the presence of DCC and HOBT in DMF followed, and the free tripeptidyl trifluoromethyl ketone vii was effectively isolated after treatment with dioxane previously saturated with hydrogen chloride gas. This ketone was obtained as a mixture of diastereomers. Isomer separation was not performed, because of a noted rapid epimerization observed in vitro as well as in vivo, most probably due to the propensity of the trifluoromethyl ketone to enolize and exist in a hydrated form.18,19 The suitable antioxidant functionalities were introduced through the use of different carboxylic acids 6, which were either commercially available or prepared by standard methods according to Scheme 3. One of these acids (6d) necessitated the BOP, DIEA peptidic coupling methodology to react with tripeptide vii-b yielding inhibitor 7 (Scheme 4). However, the majority of the carboxylic acids 6 were reacted with sulfonamides 8a,b via standard peptidic coupling conditions (DCC, DMAP) in DMF, to give esters of general formula 9 as illustrated in Scheme 4.

Chemistry Nonnatural amino acids H-Phi-OH (Phi) (4) and H-Abo-OH (Abo) (5), containing a perhydroindole and an azabicyclo[2.2.2]octane ring, respectively, were prepared according to published procedures.14-16

The compounds described in Tables 1 and 2 were prepared according to Schemes 2-5. The synthesis of the tripeptidic moiety is described in Scheme 2 and relies essentially on previously reported procedures: amino trifluoro alcohol ii was obtained in five steps from N-benzoylvaline as described by Peet.17 Compound ii was conveniently reacted with either one of our Bocprotected amino acids i using mixed anhydride coupling

Ester 9c was obtained as a secondary product during the preparation of ester 9b. Both were used directly as a mixture and separated at the next step using reverse phase HPLC. Esters of generic formula 9 were converted to the corresponding carboxylic acids 10 in the presence of NaOH in EtOH followed by acidic workup (Scheme 4). In the case of X being hydrogen on generic formula 10 and the carboxylic acid function being para with respect to the sulfonamide (compound 8a), compounds 10 were coupled directly with tripeptidic trifluoromethyl ketones viia,b using the BOP, DIEA methodology to give inhibitors 11 (route 1). The latter were generally purified by column chromatography on silica gel followed by trituration with ether or pentane. In the case of X being a chlorine atom and the carboxylic

1908

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12

Portevin et al.

Scheme 3a

a Reagents: (a) CH COOtBu, 70% HClO ; (b) EtOH, H SO ; (c) NaH, MEM-Cl, DMF; (d) NaOH, EtOH; (e) tBubromoacetate, K CO ; 3 4 2 4 2 3 (f) TFA, CH2Cl2; (g) EtONa, 4-(2-chloroethyl)benzoic acid; (h) ethyl 4-bromobutyrate, K2CO3; (i) CH3COCH2COOEt, NH4OH, EtOH.

Scheme 4a

a Reagents: (a) vii-b, BOP, DIEA, DMF; (b) 8a or 8b, DCC, DMAP, DMF; (c) NaOH, EtOH; (d) 8a, DCC, DMAP, DMF; (e) vii-a, BOP, DIEA, DMF.

acid moiety being meta to the sulfonamide (compound 8b), a second coupling reaction was performed with 8a, this time under DCC, DMAP coupling conditions, to give esters 12 (Route 2). Saponification followed by coupling with the tripeptidic entity vii-a as described above provided the two inhibitors 14a,b.

Compounds in which the N-terminal valine was replaced by an -substituted lysine were prepared as outlined in Scheme 5. The hydrochloride of -N-(benzyloxycarbonyl)lysine methyl ester (15) was reacted with sulfonamide A-OH, already used by others,6 to give intermediate 16. Removal of the -amino protecting group was effected by treatment with HBr in AcOH. This was followed by condensation with carboxylic acids 6a,e under classical peptidic coupling conditions (DCC, HOBT in DMF) to give intermediates 18 and 21. Saponification was performed to provide acids 19 and 22 which were condensed with dipeptidic trifluoromethyl ketone Phi-Val-CF3 (v-a) in the presence of DCC, HOBT, and triethylamine to give the desired inhibitors 20 and 23, respectively. Finally, as illustrated in Scheme 6, the commercially available L-serine methyl ester (24) was converted into its chloro analog by treatment with PCl5 in chloroform to give chloromethyl amino acid 25 as the hydrochloride salt.20 After protection of the amino group (Boc2O, Et3N), condensation with 4-hydroxy-3,5-di-tert-butylthiophenol in the presence of potassium carbonate in acetonitrile was performed to provide cysteine derivative 27. The required free amino group was obtained by deprotection with gaseous HCl in dioxane as described previously. Subsequent condensation with sulfonamidocarbonyl derivative A-OH, again using BOP, DIEA coupling methodology, gave intermediate 29. Saponification of the ester function under usual conditions was followed by coupling with dipeptide Phi-Val-CF3 (v-a) (DCC, HOBT, Et3N) to give the final derivative 31.

Dual Inhibition of HLE and Lipid Peroxidation

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12 1909

Scheme 5a

enzymatic preparation compared to their proline counterpart (in our hands, IC50 for ICI compound was 34 nM).

a Reagents: (a) BOP, DIEA, DMF; (b) HBr, CH COOH; (c) 6a 3 or 6c, DCC, HOBT, DMF; (d) NaOH, CH3OH; (e) v-a, HOBT, DCC, ET3N.

Scheme 6a

a Reagents: (a) PCl , CHCl ; (b) Boc O, Et N; (c) 4-hydroxy5 3 2 3 3,5-di-tert-butylthiophenol, K2CO3, CH3CN; (d) gaseous HCl, dioxane; (e) A-OH, BOP, DIEA, DMF; (f) NaOH, CH3OH; (g) v-a, DCC, HOBT, Et3N.

Enzyme Inhibitory Activities Previous studies in our laboratories had documented that it was possible to replace the central proline (Pro) of the tripeptidic backbone Val-Pro-Val found in ICI 200 880 (1) by our nonnatural amino acids Phi (4) and Abo (5) and obtain potent and specific HLE inhibitors.10,11 This was exemplified herein by compounds 32 and 33 which displayed similar IC50 values against an HLE

The aim of our ongoing investigations was to add a complementary activity to the strict elastase inhibition. Among several possible synergistic activities (mucolytic, antiinflammatory, etc.), we turned to the hypothesis that it could be beneficial to embody our inhibitors with antioxidant properties. As briefly mentioned by Stein et al.,21 the minimum structural sequence necessary to obtain a strong in vitro anti-HLE inhibitory potency was derived from H-Val-AA-Val-CF3: in our hands, the tripeptidic ketone vii-b (H-Val-Abo-Val-CF3) was found to be moderately active (IC50 ) 210 nM), which was in contrast to intermediate vi-b (Boc-Val-Abo-Val-CF3) (IC50 ) 32 nM), thus suggesting the possible importance of inhibitor interaction at the P3 region of the enzyme to obtain a good activity. The first inhibitor incorporating an antioxidant moiety (7) was quite atypical, mainly because the phenol ring was directly linked to the tripeptidic sequence through an alkylaryl chain (Table 1). Surprisingly this compound retained some in vitro inhibitory potency with an IC50 of 24 nM. On the other hand, the (acylamino)sulfonyl derivative 34, with an IC50 value of 24 nM, was equipotent to the reference inhibitor 1. This observation opened up the field of introducing different carboxylic acids via coupling with p-sulfonamidobenzoic acid or a substituted analog.

When carboxylic acid 6a was adjoined to the tripeptide through the p-sulfonamidophenyl moiety, the potencies of the corresponding Phi- and Abo-containing inhibitors 11a,b were in the same range of potency (38 and 30 nM, respectively) as those of the reference inhibitors. If the methylene linker was removed to give 11c, inhibition was still potent (IC50 ) 36 nM), as was also observed with the bis di-tert-butylphenol derivative 11d (IC50 ) 46 nM). The vinyl analogs 11e,f had similar potencies with an IC50 around 40 nM. The inhibitory potency was still robust when a 4-mercaptodi-tertbutylphenol group was linked to the acylsulfonamido

1910

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12

Table 1

Portevin et al.

Dual Inhibition of HLE and Lipid Peroxidation

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12 1911

Table 1 (Continued)

a Compounds gave satisfactory analyses ((0.4%) unless otherwise indicated. b ND: not determined, but satisfactory results by highresolution MS analysis were obtained. c C: found, 60.84; calcd 60.42. d S: found, 3.73; calcd, 3.00. e S: found, 7.93; calcd, 7.28. f S: found, 5.49; calcd. 6.60. g C: found, 57.86; calcd, 58.53. h H: found, 6.45; calcd, 5.95. i IC50 values are concentrations of compounds required to achieve 50% inhibition against HLE from human sputum. j IC50 values are concentrations of compounds required to achieve 50% inhibition against Fe3+ ascorbate-induced lipid peroxidation from liver microsomal preparation. k in: inactive.

moiety either by a methylene group as in 11g (IC50 ) 43 nM) or by an ethylphenyl group as in 11h (IC50 ) 40 nM). Compound 11i which embodied a 4-chloro-3sulfonamidobenzoic acid entity also retained a strong in vitro activity, having an IC50 value of 46 nM. At this point, we selected the trolox group22 as well as a 4-phenyldihydropyridyl entity as potential antioxidant functions, giving rise to inhibitors 11j (IC50 ) 20 nM) and 11k (IC50 ) 41 nM), respectively. It has been hypothesized that the sulfonamide functionality confers high in vitro and in vivo activity, probably through improved interactions with the S4-S5 enzymatic sites and also due to the acidic nature of this acylsulfonamide function which could prevent the rapid clearance of the compounds from the lungs.23 We took this reasoning one step further by adding a second sulfonamide function to give the bis sulfonamide derivatives 14a,b. Both compounds displayed potent in vitro inhibitory activity against HLE with IC50s of 26 and 34 nM, respectively. The preparation of an HLE inhibitor in which the P3 valine is replaced by a lysine has been documented in 1990 by Peet et al.17 This compound was reported to have a Ki value of 0.58 nM on an HLE enzymatic preparation. As illustrated in Table 2, inhibitors 20 and 23 incorporated the tripeptidic trifluoromethyl ketone Lys-Phi-Val-CF3. On the R-amino group of the lysine residue was appended the p-[p-Cl(C6H4)SO2NHCO](C6H4)CO system (A), while the -amino group on the lysine side chain was derivatized with two different 4-mercaptodi-tert-butylphenol-containing carboxylic

acids. These two compounds appeared to be slightly less potent against HLE than their related valine-containing counterparts, displaying IC50s around 70-80 nM. Finally, the P3 valine was replaced by a S-substituted cysteine to give compound 31 which presented a potency similar to that of the reference inhibitors (IC50 ) 59 nM). From these preliminary in vitro studies, it appears that (1) Phi and Abo are effective proline mimics in order to obtain potent HLE inhibitors in vitro. (2) The [(p-chlorophenyl)sulfonamido]carbonyl moiety used in the standard inhibitors can effectively be replaced by entities in which a chemical entity able to bring some additionnal antioxidant properties is incorporated: all the compounds tested presented in vitro anti-HLE potencies between 10 and 100 nM, indicating that the P4-P5 region of the enzyme has a high degree of tolerance with regard to the substituent group. Selectivity Studies Selectivity of inhibitor 14a was studied on a panel of 22 receptors24 where Ki values were higher than 10-5 M. Affinities for sodium (Veratridine) and calcium (Nifedipine) channels were superior to 10-4 and 2.4 × 10-6 M, respectively. Moreover, compound 14a gave an IC50 of 1 µM against pig pancreatic elastase (PPE) and 2 µM against prolyl endopeptidase. No inhibitory activity was detected up to 33 µM on several factors of the coagulation and the fibrinolysis cascades, including thrombin, plasmin, factor Xa, kallikrein, activated protein C, urokinase, tPa, and trypsin, as well as on

1912

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12

Portevin et al.

Table 2

a Compounds gave satisfactory analyses ((0.4%) unless otherwise indicated. b The four compounds were amorphous solids. c ND: not determined, but satisfactory results by high-resolution MS analysis were obtained. d C: found, 59.37; calcd, 60.26. N: found, 6.43; calcd, 6.89. e IC50 values are concentrations of compounds required to achieve 50% inhibition against HLE from human sputum. f IC50 values are concentrations of compounds required to achieve 50% inhibition against Fe3+ ascorbate-induced lipid peroxidation from liver microsomal preparation. g in: inactive.

acetylcholinesterase and cathepsin G. Finally, matrix metalloproteinases 1, 2, 3, and 9 were inhibited by compound 14a with IC50 values of 58, 19, 38, and 24 µM, respectively. Antioxidant Properties In vitro antioxidant potential of our inhibitors was estimated by inhibition of the Fe3+ ascorbate-induced lipid peroxidation on a rat liver microsomal preparation. As expected, ICI 200 880 was completely devoid of activity in this assay. Inhibitor 11k containing a dihydropyridine entity was virtually inactive on the peroxidation test. In contrast, the trolox group containing inhibitor 11j gave potent inhibition with an IC50 of 1.2-2.5 µM. However, for the di-tert-butylphenol moietycontaining compounds, we found that the antioxidant potency depended on the link between the phenol ring and the acylsulfonamido group; as expected, the protected phenol 11e was completely devoid of activity, and an ester function such as in 11d was also deleterious for activity. Middle range potency was obtained with direct bonding (11c, 10-20 µM), a methylene spacer (11a,b or 14b, 2.5-10 µM), or the vinylogous 11f (2.5-5 µM). The strongest inhibition of peroxidation was obtained when the 4-mercaptodi-tert-butylphenol was used: it gave rise to IC50 values between 2.5 and 5 µM for inhibitors 20, 23, and 31. Potency was further increased for inhibitors 7 and 11g,h (IC50 ) 1.2-2.5 µM) and culminated with compounds 11i and 14a with IC50s ranging between 0.6 and 1.2 µM. Furthermore, compound 14a protected the purified human LDL from copper sulfate-induced oxidative modifications with an IC50 of 0.38 µM. In our hands,

standard antioxidant compounds probucol and BHT gave IC50s of 3 µM in this assay. In Vivo Studies The main goal of our investigations was to identify compounds affording a strong and long lasting in vivo activity after intratracheal administration. The animal model used along these studies was based on the fact that HLE induces an acute hemorrhage in the hamster lung when administered intratracheally (it).25 This hemorrhage can be quantitated 3 h later by evaluation of the hemoglobin content in bronchial alveolar lavage fluid. All the compounds tested were administered it at various times prior to the HLE challenge at the dose of 15 nmol/animal. The results of this time-effect study are disclosed in Table 3. In this assay, ICI 200 880 exhibited a potent activity, since 78% of the HLE-induced lung injury was inhibited at 3 h; this activity was maintained for at least 24 h. Several novel inhibitors with strong in vitro potencies were partially or even totally devoid of in vivo activity: compound 11j in which the trolox group is appended on the (acylamino)sulfonyl linker showed no meaningful in vivo activity, as well as the bis di-tert-butylphenol derivative 11d. As expected, compound 7 exhibited a modest activity when administered 3 h before the inflammatory challenge, probably due to the lack of acylsulfonamide function within its structure; more curiously, its extented analog 11h was also poorly active. Inhibitor 11k gave a good inhibition of the HLEinduced edema at 3 h but was devoid of activity at 6 h. More generally, incorporation of the (acylamino)sulfonyl functionality within the structure of the inhibitors

Dual Inhibition of HLE and Lipid Peroxidation

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12 1913

Table 3. % inhibition of HLE-induced hemorrhage in the hamster lung at compd no.

3h

6h

18 h

24 h

48 h

ICI 200 880 7 11a 11b 11c 11d 11e 11f 11g 11h 11i 11j 11k 14a 14b 20 23 31

78 24 66 28 53 0 19 63 60 20 41 0 56 68 41 40 3 42

63

47

37

18

65 0 71

26

3

40

11

58 60

49 11

0 0

52

23

0

0 66 40 41

65 52 45

76 62 14

49

12

6

72 h

0

73 0 0

50

afforded, as anticipated, potent compounds in vivo. For example, compound 11a was found to be very potent on the reduction of the HLE-induced hemorrhage when predosed 3 and 6 h before injection of elastase; however, its activity decreased when it was administered 18 h before HLE, and no activity subsisted when administered 24 h before HLE. Noteworthy, replacement of Phi by Abo (11b) in the tripeptidic sequence resulted in almost total loss of activity. This could perhaps be explained by a difference in lipophilicity between the two central amino acids (log P(Phi) ) -1.19, log P(Abo) ) -1.75), hence between the two corresponding inhibitors, indicating that lipophilicity could also be of importance for absorption or retention of the compounds in the lungs. Compounds 11c,f,g,i presented an activity profile comparable to that of 11a with more or less potency remaining at 18 h but without any activity left when dosed at 24 or 48 h before challenge. (Note that the protected analog of 11f, 11e, was very poorly active at 3 h.) All these compounds have in common the presence of two aryl rings linked together by an X(acylamino)sulfonyl group, X being respectively a direct bond, CHdCH, CH2, or SCH2. A similar profile of activity was observed with the bis acylsulfonamide containing inhibitor 14b. More interestingly, analog 14a, with only one more sulfur atom on the phenol ring, gave a strong and very long lasting in vivo potency: since at a fixed dose of 15 nmol it, the first administration time giving 50% protection against elastase-induced lung injury was 72 h for 14a and only 18 h for ICI compound. As mentioned earlier, this long duration of action in the lung could at least in part be explained by the inability of the inhibitors to be cleared from the lung, although this was not investigated.26 This property seems to be conferred to the molecule by the presence of the acylsulfonamido moiety and is probably exacerbated if two such groups are present, as in 14a. The longer duration of action observed with 14a as compared to 14b can probably be rationalized in terms of differences in their in vivo metabolism. Finally, the lysine and cysteine derivative-containing inhibitors 20 and 31 diplayed similar potencies as inhibitors 11c,f suggesting that in vitro as well as in vivo activity can be obtained with compounds where the N-terminal valine of the tripeptidic sequence has been

Figure 1. Effects of ICI 200 880 and 14a on 14-day elastaseinduced emphysema in the hamster. C ) no treatment. PPE ) 60 IU + 0.2 mL of saline of pig pancreatic elastase. ICI 200 880 and 14a were administered it at 100 nmol in a 0.1 mL volume of saline, 3 h prior to PPE challenge.

replaced by another substituted polar amino acid. However, the length of the chain appended on the -nitrogen of the lysine residue seemed to be of prime importance since in vivo activity was completely lost when a CH2 chain (as in 20) was replaced by a (CH2)3 chain (as in 23). Furthermore, ICI 200 880 and compound 14a were compared in a model of elastase-induced chronic lung injury in the hamster, where instillation of PPE caused microvascular damage and subsequent alveolar space enlargment, i.e. emphysema.27 This was determined by quantifying the number of alveolar intercepts (mean linear intercept) on lung histological sections 14 days after the challenge (Figure 1). Intratracheal administration of ICI 200 880 just before HLE at the single dose of 100 nmol had no significant action on the increase of the alveolar spaces. In contrast, 14a significantly inhibited pulmonary lesions associated with increased alveolar spaces. Discussion and Summary Among the many HLE inhibitors described in the recent literature, the tripeptidic trifluoromethyl ketone derivatives remain one of the most promising series for aerosol administration in man. We decided to extend the field of structural variations in this class of inhibitors, and several conclusions can be drawn from the foregoing results: (1) Our proline analog amino acids have proven to be useful tools in order to obtain potent inhibitors as active as the reference compound ICI 200 880 in several assays of in vitro and in vivo reduction of HLE activity. (2) We then hypothesized that it could be worthwhile to add to the strict antielastase property of our compounds a second activity, if possible acting synergistically. We decided to embody our compounds with antioxidant properties, since several studies had recently demonstrated that the destructive potential of HLE in the lung was enhanced by active ORS. This hypothesis was chemically expressed with the use of antioxidant groups such as the tert-butylphenol ring, which was appended in different ways on the modified tripeptidic sequence. This gave rise to inhibitors that were effective HLE inhibitors in vitro with IC50 values in the 10-100 nM range. Moreover, these compounds exhibited a real antioxidant potential, since they inhib-

1914

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12

Portevin et al.

ited lipid peroxidation on a rat liver microsomal preparation as well as on purified human LDL with an IC50 similar to or even better than IC50s of reference antioxidant compounds. The elastase-induced acute lung injury was very discriminant for the selection of a leader compound in terms of duration of action in the lung: one of the most active compounds of the series, 14a, showed significant activity in this assay, reducing the hemorrhage by 50% when predosed 72 h prior to HLE challenge. Furthermore, in a 14-day model of elastaseinduced emphysema in the hamster, 14a significantly inhibited alveolar space enlargement, when administered once just before instillation of HLE. In this model, ICI 200 880 was devoid of significant activity, indicating that the addition of an antioxidant moiety in the structure of an HLE inhibitor could be relevant in order to gain some in vivo potency in a chronic model of lung injury. Although promising, these results warrant some more in vivo confirmation: to adress the importance of the antioxidant effect of our compounds, it would be interesting to compare 14a with a close analog devoid of antioxidant properties on an animal model of emphysema involving both elastase and reactive oxygen species.

layer was washed with a saturated sodium bicarbonate solution and then water. After drying on calcium sulfate and evaporation of the solvent, 15.1 g (96%) of an amorphous solid was obtained: mp 110 °C; IR (Nujol) 1764, 1680 cm-1; 1H NMR (DMSO-d6) δ 0.95 (6H, m), 1.3-2.0 (9H, m), 1.35 (9H, s), 2.3 (3H, m), 3.75 (1H, m), 4.2 (1H, m), 4.4 (1H, m), 8.4 (1H, br s). Boc-Abo-Val-CF3 (iv-b): IR (Nujol) 3311, 1761, 1670 cm-1; 1H NMR (DMSO-d ) δ 0.95 (6H, m), 1.35 (9H, s), 1.4-1.75 (7H, 6 m), 1.9 (1H, m), 2.1 (1H, m), 2.2 (1H, m), 3.9 (1H, m), 4.15 (1H, dd), 4.6-4.7 (1H, 2m). Phi-Val-CF3 (v-a). General Method 1. The protected dipeptide obtained above (15.1 g, 0.036 mol) was dissolved in 350 mL of dioxane. Gaseous HCl was bubbled at 10 °C until saturation of the solution was obtained. Stirring was continued overnight at room temperature. After evaporation of the solvent, the crude solid was taken up with pentane to give 12 g of the desired hydrochloride (94%): mp 134 °C; IR (Nujol) 3600-2400, 1764, 1680, 1171 cm-1; 1H NMR (DMSO-d6) δ 0.81.0 (6H, 2d), 1.2-1.9 (9H, m), 2.3 (3H, m), 3.6 (1H, m), 4.15 (1H, m), 4.4 (1H, m), 8.4 (1H, br s), 10.5 (1H, br s). Abo-Val-CF3 (v-b): mp 165-170 °C sublimation; IR (Nujol) 3600-2400, 1765, 1677, 1173 cm-1; 1H NMR (DMSO-d6) δ 0.81.0 (6H, 2d), 1.3-2.0 (8H, m), 2.2 (1H, m), 2.3 (1H, m), 3.4 (1H, br s), 4.0-4.2 (2H, m), 8.5 (1H, br s), 9.8 (1H, br s). Boc-Val-Phi-Val-CF3 (vi-a). General Method 2. To 6 g (0.0168 mol) of Phi-Val-CF3 in 5 mL of DMF were added 2.34 mL (0.0168 mol) of triethylamine, 3.65 g (0.0168 mol) of BocVal-OH in 3 mL of DMF, 2.57 g of HOBT (0.0168 mol) in 2 mL of DMF, and 3.46 g (0.0168 mol) of DCC in 3 mL of DMF. Stirring was continued for 20 h. DCU was filtered and DMF evaporated. The gummy residue was taken up with ethyl acetate; the organic solution was washed consecutively with brine, a saturated sodium bicarbonate solution, brine, a 10% citric acid solution, brine, and water. Solution was dried over calcium sulfate, filtered, and evaporated to give 9.1 g of a crude solid. This product was purified by chromatography (CH2Cl2EtOH, 98:2) to give 4.4 g of the protected tripeptide: IR (Nujol) 3276, 1765, 1687, 1207-1150 cm-1; 1H NMR (DMSO-d6) δ 0.81.0 (12H, m), 1.15-1.9 (8H, m), 1.4 (9H, s), 1.9 (1H, m), 2.2 (2H, m), 3.75 (1H, m), 4.0-4.4 (2H, m), 4.45-4.65 (1H, 2m). Boc-Val-Abo-Val-CF3 (vi-b): IR (Nujol) 3500-3300, 1793, 1697, 1624 cm-1; 1H NMR (CDCl3) δ 0.8-1.1 (12H, m), 1.4 (9H, s), 1.5-1.7 (4H, m), 1.8-2.0 (4H, m), 2.4 (1H, m), 2.6 (2H, m), 3.95 (1H, m), 4.4 (1H, m), 5.02 (2H, m), 6.45 (1H, br s), 6.7 (1H, br s). Val-Phi-Val-CF3 (vii-a): see General Method 1; IR (Nujol) 1763, 1695 cm-1; 1H NMR (DMSO-d6) δ 0.7-1.1 (12H, m), 1.12.5 (13H, m), 3.5-4.8 (4H, m). Val-Abo-Val-CF3 (vii-b): IR (Nujol) 1761, 1693, 1631 cm-1; 1H NMR (DMSO-d ) δ 1.0 (12H, m), 1.3-1.8 (8H, m), 1.8-2.2 6 (3H, m), 4.0 (1H, m), 4.1 (1H, m), 4.4 (1H, dt), 4.6 (1H, 2m). 4-[2-(4-Hydroxy-3,5-di-tert-butylphenyl)thio]ethyl]benzoyl-(S)-Val-(S)-Abo-(R,S)-Val-CF3 (7). General Method 3. Tripeptide vii-b (662 mg, 0.015 mol), carboxylic acid 6d (580 mg, 0.015 mol), BOP (668 mg, 0.015 mol), and DIEA (0.79 mL, 0.045 mol) were stirred at room temperature for 18 h in 50 mL of DMF. After evaporation of the solvent, the residue was taken up with 100 mL of ethyl acetate. The solution was washed with water, dried over calcium sulfate, and evaporated. Purification by chromatography (CH2Cl2-AcOEt, 90:10) gave 0.7 g (60%) of the title compound: IR (Nujol) 1761, 1640-1700 cm-1; 1H NMR (DMSO-d6) δ 0.95 (12H, m), 1.4 (18H, s), 1.52.0 (9H, m), 2.15 (2H, m), 2.9 (2H, t), 3.1 (2H, t), 4.1-4.7 (4H, m), 7.15 (2H, s), 7.25 (2H, d), 7.75 (2H, d). Ethyl 4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]benzoate (9a). General Method 4. To a solution of 5 g (0.019 mol) of 4-hydroxy-3,5-di-tert-butylphenylacetic acid (6a) in 60 mL of DMF were successively added 4.35 g (0.019 mol) of ethyl 4-sulfonamidobenzoate (8a) in 10 mL of DMF, 2.32 g (0.019 mol) of DMAP in 10 mL of DMF, and DCC (3.92 g, 0.019 mol). The resulting solution was stirred at room temperature for 2 days. DCU was filtered and DMF was concentrated under reduced pressure. The residue was treated with ethyl acetate. The organic layer was washed consecutively with saturated aqueous NaCl, 10% aqueous citric acid, and again brine. The EtOAc was dried (CaSO4) and

Experimental Section Melting points were determined on a Tottoli apparatus and were not corrected. Elemental analyses were carried out by the analytical department of the Institut de Recherches Servier; results obtained for specified elements are within (0.4% of the theoretical values. IR spectra were recorded on a Bruker IFS 28 spectrophotometer. 1H NMR spectra of deuteriochloroform or DMSO-d6 solutions were recorded on a Bruker AC 200 or AM 300 spectrometer. Chemical shifts are given in ppm with TMS as the internal standard. Optical rotations were recorded with a 241 Perkin Elmer polarimeter. Abbreviations: DCC ) N, N ′-dicyclohexylcarbodiimide; DCU ) N, N ′-dicyclohexylurea; HOBT ) 1-hydroxybenzotriazole; BOP ) (benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophosphate; DMAP ) 4-(dimethylamino)pyridine; DIEA ) N,N-diisopropylethylamine; NMM ) N-methylmorpholine. (2S,3aS,7aS)-1-(tert-Butyloxycarbonyl)-2-[[(1-isopropyl2-hydroxy-3,3,3-trifluoropropyl)amino]carbonyl]perhydroindole (iii-a). To a mixture of 12.4 g (0.046 mol) of Boc-Phi-OH (i)9-11 and 5 mL (0.046 mol) of N-methylmorpholine in 80 mL of THF was added dropwise at -10 °C a solution of 5.95 mL (0.046 mol) of isobutyl chloroformate in 10 mL of THF. After 10 min at -10 °C, a solution of the amino alcohol ii (9.5 g, 0.046 mol) and N-methylmorpholine (5 mL) in 60 mL of DMF was added dropwise at -10 °C. Stirring was continued overnight and temperature slowly raised to ambient. Solvents were evaporated, and the residue was taken up with isopropyl ether (150 mL); after filtration of an insoluble, the organic phase was washed with water and dried over calcium sulfate. Evaporation of the solvent was followed by trituration with pentane to give 16.8 g (83%) of the desired intermediate: mp 193 °C; IR (Nujol) 3600-3300, 1662 cm-1; 1H NMR (DMSO-d ) δ 0.95 (6H, m), 1.3-2.0 (9H, m), 1.4 (9H, 6 s), 2.3 (3H, m), 3.75-4.4 (4H, m), 6.5 (1H, br s), 8.4 (1H, br s). (3S)-2-Aza-2-(tert-butyloxycarbonyl)-3-[[(1-isopropyl2-hydroxy-3,3,3-trifluoropropyl)amino]carbonyl]bicyclo[2.2.2]octane (iii-b): IR (Nujol) 3600-3300, 1662, 1165, 1138, 1105 cm-1; 1H NMR (DMSO-d6) δ 0.8 (6H, m), 1.35-2.0 (9H, m), 1.4 (9H, s), 2.15 (1H, m), 3.7-4.15 (4H, m), 6.35 (1H, br s), 7.2 and 7.6 (1H, br s). Boc-Phi-Val-CF3 (iv-a). Alcohol obtained at the previous step (15.8 g, 0.037 mol) was oxidized by the DMSO-Ac2O mixture (160 and 130 mL, respectively). After stirring at room temperature during 48 h, 300 mL of water was added; stirring was continued for 3 more h and the resulting solid filtrated. This solid was taken up with dichloromethane; the organic

Dual Inhibition of HLE and Lipid Peroxidation

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12 1915

concentrated in vacuo to give the crude title compound as a dark solid. This compound was purified by chromatography over silica gel (CH2Cl2-CH3OH-NH4OH, 90:10:1) to give a colorless oil (6.8 g, 75.5%) that crystallized upon trituration with pentane: mp 150 °C; IR (Nujol) 3600, 3300-3000, 1726, 1711-1685, 1360, 1144 cm-1; 1H NMR (DMSO-d6) δ 1.3 (21H, t + s), 3.4 (2H, s), 4.35 (2H, q), 6.85 (2H, s), 6.9 (1H, br s), 8.0 (2H, d), 8.15 (2H, d), 12.45 (1H, br s). Ethyl 4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)carbonyl]amino]sulfonyl]benzoate (9b) and Ethyl 4-[[[[4-[(4Hydroxy-3,5-di-tert-butylphenyl)carboxy]-3,5-di-tert-butylphenyl]carbonyl]amino]sulfonyl]benzoate (9c). To a 100 mL toluene solution of the sulfonamide 3 (3.45 g, 0.015 mol) was added 2.1 mL (0.015 mol) of triethylamine; then at 80 °C the acid chloride of 4-hydroxy-3,5-di-tert-butylbenzoic acid (4.05 g, 0.015 mol) (itself prepared with 2 mol equiv of oxalyl chloride in toluene at 80 °C) was added. The reaction mixture was refluxed overnight and cooled down to room temperature, and water was added. The remaining starting sulfonamide was filtered, and the organic layer was washed several times with water. Drying (CaSO4) and evaporation gave an amorphous solid (5.5 g). This compound was purified by chromatography (CH2Cl2-CH3OH-NH4OH, 90:10:0.5) to give 3.3 g of a mixture of both 9b,c. Ethyl 4-[[[4-(ethoxymethoxy)-3,5-di-tert-butylcinnamoyl]amino]sulfonyl]benzoate (9d): IR (Nujol) 3217, 1726, 1622, 1275, 1109 cm-1; 1H NMR (DMSO-d6) δ 1.2 (3H, t), 1.35 (3h, t), 1.4 (18H, s), 3.8 (2H, q), 4.35 (2H, q), 4.9 (2H, s), 6.55 (1H, d), 7.5 (1H, d), 7.45 (2H, s), 8.15 (4H, dd), 12.3 (1H, br s). Ethyl 4-[[[[(4-hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]benzoate (9e): mp 188 °C; IR (Nujol) 3600, 3245, 1722, 1699 cm-1; 1H NMR (DMSO-d6) δ 1.3 (18H, s), 1.35 (3H, t), 3.55 (2H, s), 4.35 (2H, q), 6.95 (2H, s), 7.10 (1H, br s), 7.55 (1H, br s), 8.00-8.2 (4H, m). Ethyl 3-[[[[(4-hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]-4-chlorobenzoate (9g): IR (Nujol) 3500-3000, 1718, 1593 cm-1; 1H NMR (DMSO-d6) δ 1.3 (21H, m), 3.35 (2H, s), 4.35 (2H, q), 6.9 (1H, s), 7.05 (2H, s), 7.55 (1H, d), 7.95 (1H, dd), 8.5 (1H, d). Ethyl 3-[[[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]-4-chlorobenzoate (9h): IR (Nujol) 36103583, 3097, 1726, 1684 cm-1; 1H NMR (CDCl3) δ 1.45 (3H, t), 1.45 (18H, s), 3.55 (2H, s), 4.4 (2H, q), 5.3 (1H, br s), 6.95 (2H, s), 7.55 (1H, d), 8.1 (1H, br s), 8.2 (1H, dd), 8.9 (1H, d). Ethyl 4-[[[(2(R,S)-6-hydroxy-2,5,7,8-tetramethylchroman-2-yl)carbonyl]amino]sulfonyl]benzoate (9i): IR (Nujol) 3700-2500, 1707, 1649, 1375, 1141 cm-1; 1H NMR (CH3OH-d4) δ 1.4 (3H, t), 1.45 (3H, s), 1.65 (1H, m), 1.9-2.15 (9H, 3s), 2.35 (3H, m), 4.35 (2H, q), 7.65 (2H, d), 7.9 (2H, d). Ethyl 4-[[[(2,6-dimethyl-3,5-dicarbethoxy-1,4-dihydropyridin-4-yl)carbonyl]amino]sulfonyl]benzoate (9j): mp 202 °C; IR (Nujol) 3500-3000, 1737-1676, 1136, 1130 cm-1; 1H NMR (DMSO-d ) δ 1.1 (6H, t), 1.3 (3H, t), 2.25 (6H, s), 4.0 6 (4H, m), 4.3 (2H, q), 4.85 (1H, s), 7.05 (2H, d), 7.7 (2H, d), 7.9 (4H, m), 8.75 (1H, br s). 4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]benzoic Acid (10a). General Method 5. A 5 g (0.0105 mol) portion of ester 9a was dissolved in 125 mL of ethanol. Nitrogen was bubbled through the solution for 1 h; then 22 mL (0.021 mol) of 1 N NaOH was added. The reaction mixture was stirred at room temperature for 2 h, followed by addition of 25 mL of water. The reaction mixture was then heated at 60 °C for 20 h under nitrogen. After cooling, the solvents were evaporated under reduced pressure, and the residue was taken up with water. The aqueous layer was extracted with ethyl acetate, acidified with 1 N HCl (30 mL), and extracted with ethyl acetate. The organic layer was washed with water to neutral pH, dried over calcium sulfate, and evaporated to give 4.2 g of a crude solid. This material was crystallized from pentane to give 4 g (70%) of the title compound: mp 228 °C; IR (Nujol) 3635, 3300-2400, 1707 cm-1; 1H NMR (DMSO-d ) 1.3 (18H, s), 3.4 (2H, s), 6.85 (2H, s), 7.5 6 (1H, br s), 7.95 (2H, d), 8.1 (2H, d), 12.4 (1H, br s), 13.3 (1H, br s). 4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)carbonyl]amino]sulfonyl]benzoic Acid (10b) and 4-[[[[4-[(4-Hydroxy-

3,5-di-tert-butylphenyl)carboxy]-3,5-di-tert-butylphenyl]carbonyl]amino]sulfonyl]benzoic Acid (10c). Separation of the two compounds (on a 2 g scale) was performed by reverse phase HPLC on a RP 18 15/25 column (5 mL/min) with CH3CN-H2O-CH3COOH (70:30:1) as eluent. After evaporation of the eluent, the crude powder was triturated in pentane to give 450 mg of 10b and 700 mg of 10c: mp 272 °C; IR (Nujol) 3626, 3298, 3000-2400, 1685 cm-1; 1H NMR (DMSO-d6) δ 1.35 (18H, s), 7.2 (1H, br s), 7.5 (1H, br s), 7.7 (2H, s), 7.95 (4H, m), 12.7 (1H, br s); IR (Nujol) 3570, 3257, 1716, 1685 cm-1; 1H NMR (DMSO-d6) δ 1.25 (18H, s), 1.4 (18H, s), 7.8 (2H, s), 7.9 (2H, s), 8.15 (4H, m), 13.5 (1H, br s). 4-[[[4-(Ethoxymethoxy)-3,5-di-tert-butylcinnamoyl]amino]sulfonyl]benzoic acid (10d): IR (Nujol) 3101, 1687, 1612 cm-1; 1H NMR (DMSO-d6) δ 1.2 (3H, t), 1.4 (18H, s), 3.8 (2H, q), 4.9 (2H, s), 6.55 (1H, d), 7.5 (2H, s), 7.55 (1H, d), 8.15 (4H, dd). 4-[[[[(4-Hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]benzoic acid (10e): mp 240°C; IR (Nujol) 3607, 3279, 3200-2450, 1703 cm-1; 1H NMR (DMSO-d6) δ 1.3 (18H, s), 3.55 (2H, s), 7.0 (2H, s), 7.1 (1H, br s), 7.95-8.2 (4H, 2d), 13-13.5 (1H, br s). 4-[[[4-[2-[(4-Hydroxy-3,5-di-tert-butylphenyl)thio]ethyl]benzoyl]amino]sulfonyl]benzoic acid (10f): IR (Nujol) 3610, 3500-2400, 1703, 1360, 1171 cm-1; 1H NMR (DMSOd6) δ 1.35 (18H, s), 2.85 (2H, t), 3.15 (2H, t), 7.0 (1H, br s), 7.05 (2H, s), 7.35 (2H, d), 7.8 (2H, d), 8.15 (4H, m). 3-[[[[(4-Hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]-4-chlorobenzoic acid (10g): IR (Nujol) 3600, 3500-3000, 1709 cm-1; 1H NMR (DMSO-d6) δ 1.3 (18H, s), 3.55 (2H, s), 7.0 (2H, s), 7.1 (1H, s), 7.8 (1H, d), 8.15 (1H, dd), 8.55 (1H, d), 12.5 (1H, br s). 3-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]-4-chlorobenzoic acid (10h): IR (Nujol) 3641, 3350-2500, 1709, 1682, 1377, 1109 cm-1; 1H NMR (DMSOd6) δ 1.3 (18H, s), 3.45 (2H, s), 6.8 (1H, br s), 6.85 (2H, s), 7.75 (1H, d), 8.15 (1H, dd), 8.55 (1H, d), 12.8 (1H, br s), 13.6 (1H, br s). 4-[[[(2(R,S)-6-Hydroxy-2,5,7,8-tetramethylchroman-2yl)carbonyl]amino]sulfonyl]benzoic acid (10i): IR (Nujol) 3650-2300, 1724-1700, 1354, 1184 cm-1; 1H NMR (DMSOd6) δ 1.35 (3H, s), 1.65 (1H, m), 1.85 (3H, s), 2.0-2.4 (3H, m), 2.1 (6H, 2s), 7.5 (1H, br s), 7.80 (2H, d), 8.05 (2H, d), 11.9 (1H, br s), 13.5 (1H, br s). 4-[[[(2,6-Dimethyl-3,5-dicarbethoxy-1,4-dihydropyridin4-yl)carbonyl]amino]sulfonyl]benzoic acid (10j): IR (Nujol) 3600-2500, 1685, 1655, 1336, 1128 cm-1; 1H NMR (DMSOd6) δ 1.1 (6H, m), 2.25 (6H, s), 3.95 (4H, m), 4.85 (1H, s), 7.05 (2H, d), 7.7 (2H, d), 7.75-7.9 (4H, 2d), 8.8 (1H, br s). Ethyl 4-[[[[4-chloro-3-[[[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]phenyl]carbonyl]amino]sulfonyl]benzoate (12b): IR (Nujol) 3627, 3232, 1705 cm-1; 1H NMR (DMSO-d ) δ 1.3 (3H, t), 1.3 (18H, s), 3.45 (2H, s), 6 4.35 (2H, q), 6.8 (1H, br s), 6.85 (2H, s), 7.7 (1H, d), 8.1 (5H, m), 8.5 (1H, d), 12.8 (1H, br s). 4-[[[[4-Chloro-3-[[[[(4-hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]phenyl]carbonyl]amino]sulfonyl]benzoic acid (13a): IR (Nujol) 3500-3000, 35002500, 1705 cm-1; 1H NMR (DMSO-d6) δ 1.25 (18H, s), 3.55 (2H, s), 7.0 (2H, s), 7.1 (1H, br s), 7.75 (1H, s), 7.8 (1H, d), 8.15 (5H, m), 8.55 (1H, d), 12.8 (1H, br s). 4-[[[[4-Chloro-3-[[[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]phenyl]carbonyl]amino]sulfonyl]benzoic acid (13b): IR (Nujol) 3627-3221, 2600, 1703 cm-1; 1H NMR (DMSO-d ) δ 1.3 (18H, s), 3.5 (2H, s), 6.9 (2H, 2s), 6 7.8 (1H, d), 8.15 (5H, m), 8.5 (1H, d), 12.9 (1H, br s). 4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-ValCF3 (11a): prepared according to General Method 3 from 10a and tripeptide vii-a; IR (Nujol) 3639, 3400-3100, 1763-1650, 1360, 1673 cm-1; 1H NMR (DMSO-d6) δ 0.8-0.95 (12H, m), 1.2-1.8 (8H, m), 1.3 (18H, s), 1.8-2.0 (2H, m), 2.15-2.3 (3H, m), 3.35 (2H, s), 4.2-4.5 (3H, m), 4.7 (1H, m), 6.8 (1H, br s), 6.85 (2H, s), 7.9-8.05 (4H, d), 8.5-8.7 (1H, 2d), 8.9 (1H, br s), 12.3 (1H, br s).

1916

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12

Portevin et al.

4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]benzoyl-(S)-Val-(S)-Abo-(R,S)-Val-CF3 (11b): IR (Nujol) 3700-2500, 1761, 1700-1620, 1370, 1173 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.3 (9H, s), 1.4-1.8 (8H, m), 2.0-2.25 (3H, m), 3.4 (2H, s), 4.0-4.7 (4H, m), 6.8 (1H, br s), 6.85 (2H, s), 7.5 (1H, d), 7.9 (2H, d), 8.05 (2H, d), 8.4 (1H, d), 8.7 (1H, d), 12.4 (1H, br s). 4-[[[(4-Hydroxy-3,5-di-tert-butylphenyl)carbonyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-ValCF3 (11c): IR (Nujol) 3624, 3278, 1763-1622 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.1-2.0 (8H, m), 1.4 (18H, s), 1.95 (2H, m), 2.2 (3H, m), 4.25-4.5 (4H, m), 7.55 (2H, s), 8.0 (4H, q), 8.4 (1H, br s), 8.9 (1H, br s), 12.5 (1H, br s). 4-[[[[4-[(4-Hydroxy-3,5-di-tert-butylphenyl)carboxy]3,5-di-tert-butylphenyl]carbonyl]amino]sulfonyl]benzoyl(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-Val-CF3 (11d): IR (Nujol) 3626, 3359, 1763-1620 cm-1; 1H NMR (DMSO-d6) δ 0.95 (12H, m), 1.25 (4H, m), 1.3 (18H, s), 1.4 (18H, s), 1.7 (4H, m), 2.0 (3H, m), 2.2 (3H, m), 4.2-4.5 (3H, m), 7.8 (2H, s), 7.95 (2H, s), 8.1 (4H, dd), 8.5 (2H, br s), 8.9 (2H, br s). 4-[[[4-(Ethoxymethoxy)-3,5-di-tert-butylcinnamoyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-ValCF3 (11e): IR (Nujol) 3269, 1763, 1668-1624 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.25 (3H, t), 1.4 (18H, s), 1.7-1.9 (11H, m), 2.2 (2H, m), 3.8 (2H, q), 4.35-4.8 (4H, m), 4.9 (2H, s), 6.6-7.5 (2H, d), 7.5 (2H, s), 7.7 (1H, d), 8.1 (4H, dd), 8.8 (1H, d), 12.2 (1H, br s). 4-[[(4-Hydroxy-3,5-di-tert-butylcinnamoyl)amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-Val-CF3 (11f): prepared according to General Method 1 from 11e; IR (Nujol) 3627, 3276, 1763-1622 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.2-2.3 (13H, m), 1.4 (18H, s), 4.3-4.7 (4H, m), 6.5 (1H, d), 7.3 (2H, s), 7.5 (1H, d), 7.6 (1H, br s), 8.05 (4H, dd), 8.68.9 (1H, br s). 4-[[[[(4-Hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)Val-CF3 (11g): IR (Nujol) 3633, 3300, 1763, 1720, 1700, 1660, 1622, 1527 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.0-2.3 (13H, m), 1.80 (18H, s), 3.1 (1H, m), 3.5 (2H, s), 3.9 (3H, m), 4.7 (1H, t), 6.7 (1H, d), 7.0 (2H, s), 7.05 (1H, s), 8.0 (4H, m), 8.60 (1H, d). 4-[[[4-[2-[(4-Hydroxy-3,5-di-tert-butylphenyl)thio]ethyl]benzoyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)Phi-(R,S)-Val-CF3 (11h): IR (Nujol) 3633, 3500-3200, 1761, 1720-1650, 1350, 1159 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.2-1.7 (8H, m), 1.4 (18H, s), 1.8-2.3 (5H, m), 2.8 (2H, t), 3.1 (2H, t), 4.0-4.5 (3H, m), 4.3-4.7 (1H, 2m), 7.05 (1H, br s), 7.1 (2H, s), 7.15 (2H, s), 7.8 (2H, d), 7.9 (4H, d), 8.5-8.7 (1H, 2d). 3-[[[[(4-Hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]-4-chlorobenzoyl-(S)-Val-(2S,3aS,7aS)Phi-(R,S)-Val-CF3 (11i): IR (Nujol) 3633, 3307, 1763-1620 cm-1; 1H NMR (DMSO-d6) δ 0.9 (12H, m), 1.3 (18H, s), 1.52.3 (12H, m), 3.6 (2H, s), 4.5 (4H, m), 7.0 (2H, s), 7.1 (1H, s), 7.8 (1H, d), 8.25 (1H, d), 8.6 (1H, d). 4-[[[(2-(R,S)-6-Hydroxy-2,5,7,8-tetramethylchroman-2yl)carbonyl]amino]sulfonyl]benzoyl-(S)-Val-(S)-Abo-(R,S)Val-CF3 (11j): IR (Nujol) 3700-3000, 1761, 1730-1626, 1354, 1150 cm-1; 1H NMR (DMSO-d6) δ 0.8-1.0 (12H, m), 1.3-1.75 (8H, m), 1.35 (3H, s), 1.8-2.3 (6H, m), 1.9 (3H, s), 2.1 (6H, 2s), 2.4 (1H, m), 4.0-4.7 (4H, 4m), 6.6-6.9 and 7.1-7.4 (1H, m), 7.4 (1H, br s), 7.8-7.95 (4H, 2d), 8.3-8.6 (1H, 2d), 11.8 (1H, br s). 4-[[[(2,6-Dimethyl-3,5-dicarbethoxy-1,4-dihydropyridin4-yl)carbonyl]amino]sulfonyl]benzoyl-(S)-Val-(S)-Abo(R,S)-Val-CF3 (11k): IR (Nujol) 3330, 1730-1650, 1350, 1122 cm-1; 1H NMR (DMSO-d6) δ 0.8-1.0 (12H, n), 1.1 (6H, t), 1.21.7 (8H, m), 1.8-2.2 (3H, m), 2.25 (6H, s), 3.95 (4H, m), 4.04.4 (3H, m), 4.65 (1H, m), 4.85 (1H, s), 7.1 (2H, d), 7.3-7.5 (2H, m), 7.65 (2H, d), 7.80 (4H, m), 8.5 (1H, 2m), 8.75 (1H, s). 4-[[[[4-Chloro-3-[[[[(4-hydroxy-3,5-di-tert-butylphenyl)thio]acetyl]amino]sulfonyl]phenyl]carbonyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-ValCF3 (14a): IR (Nujol) 3340, 1761-1622 cm-1; 1H NMR (DMSOd6) δ 0.9 (12H, m), 1.2-2.3 (8H, m), 1.3 (18H, s), 3.6 (2H, s),

4.4 (4H, m), 7.0 (2H, s), 7.1 (1H, s), 7.8 (1H, d), 8.1-8.2 (5H, m), 8.6 (1H, d). 4-[[[[4-Chloro-3-[[[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]amino]sulfonyl]phenyl]carbonyl]amino]sulfonyl]benzoyl-(S)-Val-(2S,3aS,7aS)-Phi-(R,S)-Val-CF3 (14b): IR (Nujol) 3500-3000, 1763-1622, 1377, 1173 cm-1; 1H NMR (DMSO-d6) δ 0.85 (12H, m), 1.1-2.2 (8H, m), 1.3 (18H, s), 1.75 (2H, dd), 2.15 (2H, m), 2.25 (1H, m), 3.45 (2H, s), 4.15-4.7 (4H, m), 6.8 (2H, s), 7.7 (1H, d), 8.0 (5H, m), 8.1 (1H, br s), 8.45 (1H, d), 8.8 (1H, br s), 12.8 (1H, br s). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-(benzyloxycarbonyl)-(S)-Lys-OCH3, hydrochloride (16): was prepared according to General Method 3 and directly treated without any further purification. 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoyl(S)-Lys-OCH3, Hydrochloride (17). A 1.2 g (0.0019 mol) sample of the hydrochloride obtained in the previous step was dissolved in 50 mL of acetic acid. A 33% HBr in acetic acid solution (50 mL) was added, and the reaction was continued for 4 h. After evaporation of the solvent under reduced pressure, the residue was taken up with 80 mL of ethyl acetate. The organic layer was washed with a saturated sodium bicarbonate solution to neutral pH and evaporated. Treatment of the residue with dry methanol to eliminate traces of inorganic salts was followed by column chromatography (CH2Cl2-CH3OH, 60:40) to give 1 g of the deprotected lysine derivative, probably containing a small amount of inorganic salts: IR (Nujol) 3600-3163, 1736, 1645, 1333, 1132 cm-1; 1H NMR (DMSO-d6) δ 1.2-1.85 (6H, m), 2.75 (2H, t), 3.6 (3H, s), 4.45 (1H, t), 7.2 (1H, br s), 7.8 (4 + 1H, 2d + br s), 7.9 (2H, d), 8.7 (1H, br s). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]-(S)-LysOCH3 (18): prepared from carboxylic acid 6a and lysine derivative 17 described above according to General Method 2 and directly treated without any further purification. 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-[[(4-hydroxy-3,5-di-tert-butylphenyl)thio]butyroyl](S)-Lys-OCH3 (21): IR (Nujol) 3636, 3329, 1740, 1639, 1339, 1129 cm-1; 1H NMR (DMSO-d6) δ 1.3-1.8 (8H, m), 1.35 (18H, s), 2.2 (2H, t), 2.8 (2H, t), 3.05 (2H, m), 3.65 (3H, s), 4.4 (1H, t), 5.55 (1H, br s), 7.0 (1H, br s), 7.05 (2H, s), 7.45 (2H, d), 7.85 (4H, m), 7.95 (2H, d), 8.7 (1H, d). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]-(S)-LysOH (19): prepared according to General Method 5 and directly treated without any further purification. 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-[(4-hydroxy-3,5-di-tert-butylphenyl)thio]butyroyl](S)-Lys-OH (22): IR (Nujol) 1705, 1647, 1541, 1350, 1169 cm-1; 1H NMR (DMSO-d6) δ 1.35 (18H, s), 1.40 (4H, m), 1.7 (2H, q), 1.8 (2H, m), 2.15 (2H, t), 2.75 (2H, t), 3.0 (2H, q), 4.35 (1H, m), 6.95 (1H, s), 7.05 (2H, s), 7.75 (2H, d), 7.95 (4H, s), 8.05 (2H, d), 8.70 (1H, d), 12.6 (1H, br s). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-[(4-hydroxy-3,5-di-tert-butylphenyl)acetyl]-(S)-Lys(2S,3aS,7aS)-Phi-(R,S)-Val-CF3 (20): prepared according to General Method 2, using the intermediate dipeptide Phi-ValCF3 (v-a) described above; IR (Nujol) 3639, 3315, 1741, 1645 cm-1; 1H NMR (DMSO-d6) δ 1.0-2.0 (16H, m), 1.1 (6H, d), 1.4 (18H, s), 2.3 (2H, m), 3.0 (2H, m), 3.2 (2H, s), 6.6 (1H, s), 6.9 (2H, s), 7.4 (2H, d), 8.0 (6H, m). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylN-E-[[(4-hydroxy-3,5-di-tert-butylphenyl)thio]butyroyl](S)-Lys-(2S,3aS,7aS)-Phi-(R,S)-Val-CF3 (23): IR (Nujol) 3600-3000, 1639, 1541 cm-1; 1H NMR (DMSO-d6) δ 0.7-1.0 (6H, m), 1.1-1.8 (19H, m), 1.35 (18H, s), 2.15 (3H, m), 2.75 (3H, m), 3.0 (2H, m), 4.0 (1H, m), 4.40 (1H, m), 4.55 (1H, m), 7.05 (2H, s), 7.50 (2H, d), 7.85 (4H, m), 7.92 (2H, d). N-(tert-Butyloxycarbonyl)-S-(4-hydroxy-3,5-di-tert-butylphenyl)-(S)-Cys-OCH3 (27). A 8.2 g (0.039 mol) portion of R-[(tert-butyloxycarbonyl)amino]-β-chloropropionic acid (26) was reacted under nitrogen with thiophenol xii (11.9 g, 0.05 mol) and potassium carbonate (5.53 g, 0.04 mol) in dry acetonitrile (120 mL) at 65 °C for 20 h. After filtration and concentration under reduced pressure, the residue was dis-

Dual Inhibition of HLE and Lipid Peroxidation

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12 1917

solved in ethyl acetate. The organic layer was washed with water to neutral pH and then with brine. Drying (CaSO4) and evaporation gave a crude solid, which was purified by column chromatography (cyclohexane-EtOAc, 90:10). The resulting solid was triturated with pentane to give 9.4 g (55%) of the title intermediate: IR (Nujol) 3600, 3336, 1732, 1695 cm-1; 1 H NMR (DMSO-d6) δ 1.4 (18H, s), 2.95-3.2 (2H, dd), 3.6 (3H, s), 4.1 (1H, m), 7.15 (2H, s), 7.3 (1H, d). S-(4-Hydroxy-3,5-di-tert-butylphenyl)-(S)-Cys-OCH3, hydrochloride (28): prepared according to General Method 1; IR (Nujol) 3633, 3400-1980, 1751 cm-1; 1H NMR (DMSO-d6) δ 1.4 (18H, s), 3.4 (3H, s), 3.6 (2H, m), 4.4 (1H, m), 5.3 (1H, s), 7.3 (2H, s). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylS-(4-hydroxy-3,5-di-tert-butylphenyl)-(S)-Cys-OCH3 (29): prepared according to General Method 3 and directly treated without any further purification. 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylS-(4-hydroxy-3,5-di-tert-butylphenyl)-(S)-Cys-OH (30): prepared according to General Method 5; IR (Nujol) 3629, 34002500, 1707, 1653 cm-1; 1H NMR (DMSO-d6) δ 1.35 (18H, s), 3.1-3.4 (2H, m), 4.5 (1H, m), 7.1 (3H, m), 7.6-8.0 (8H, m), 8.8 (1H, d). 4-[[[(4-Chlorophenyl)sulfonyl]amino]carbonyl]benzoylS-(4-hydroxy-3,5-di-tert-butylphenyl)-(S)-Cys-(2S,3aS,7aS)-Phi-(R,S)-Val-CF3 (31): prepared according to General Method 2; IR (Nujol) 3650-2500, 1761, 1643, 1539 cm-1; 1H NMR (DMSO-d6) δ 0.6-0.95 (6H, m), 1.1-2.5 (13H, m), 1.35 (18H, 2s), 3.10 (2H, m), 4.0 (1H, m), 3.35 (1H, m), 4.5 (1H, m), 7.15 (2H, m), 7.45 (2H, d), 7.85 (4H, m), 7.90 (2H, t). Enzymatic Inhibition of HLE.28 Inhibition of HLE was assayed spectrophotometrically at 37°C by continuous monitoring of the release of p-nitroaniline at 410 nm from the substrate methoxysuccinyl-Ala-Ala-Pro-Ala-p-nitroanilide. Results are listed in Tables 1 and 2 and are expressed in terms of IC50. Lipid Peroxidation Assay: (a) Hepatic Microsomes. Rat livers were homogenized in phosphate buffer (0.01 M, pH 7.4). After differential centrifugation, microsomal proteins were used at 0.5 mg/mL for the lipid peroxidation assay and incubated with FeCl3 (100 µM) and ascorbate (10 µM) for 30 min. Lipid peroxidation was measured as thiobarbituric acidreactive substances (TBARS) by mixing 200 µL of the sample with 2 mL of a 0.3% thiobarbituric acid solution in 15% trichloroacetic acid and 0.02% tert-butylhydroxytoluene. The mixture was kept at 95°C for 25 min, and the assessment of MDA-thiobarbituric acid complex was determined by measuring the absorbance of the solution at 532 nm.29 (b) Human LDL Oxidation. LDL were prepared from fresh normal human plasma by sequential ultracentrifugation according to Havel et al.30 Oxidation was performed by incubating 0.2 mg of LDLwith 5 µM CuSO4 in 1 mL of serumfree HamF10 medium for 24 h at 37°C with or without test products. The extent of lipid peroxidation was assessed by measuring the TBARS according to the same method as described for hepatic microsomes. Elastase-Induced Hemorrhage in the Hamster Lung.25 Ten male golden Syrian hamsters 10 weeks of age, weighing 100-120 g, were used in each group. Hamsters were anesthetized by ip injection of nembutal, and the trachea were surgically exposed. Drugs solubilized in 20% DMSO were injected intratracheally (it) at a dose of 15 nmol in a 0.1 mL volume. At different times (3, 18, 24, 48, and 72 h) purified human sputum elastase (Elastin Products Co., Owensvill, MO), 50 IU in 0.2 mL of saline, was administered directly into the trachea. Three hours later, animals were killed by an overdose of nembutal, and lung lavage was performed using a single 3 mL aliquot of 0.9% saline in a 5 mL syringe by gently expanding the lungs and then withdrawing the saline a total of five times, yielding a final volume of approximately 2.5 mL of BAL (bronchoalveolar lavage) fluid from each animal. Hemorrhage was quantified by the amount of blood in each BAL sample, determined by using the Boehringer Manheim Diagnostica kit for hemoglobin. Results are expressed as

percent inhibition of pulmonary hemorrhage in treated animals as compared to control. Fourteen-Day Emphysema Model.27 Six male golden Syrian hamsters weighing 100-120 g were used in each group. Anesthesia and drug administration were performed as described above. Drugs were solubilized in water and injected it at a dose of 100 nmol in a 0.1 mL volume. Three hours later, purified pig pancreatic elastase (Elastin Products Co.), 60 IU in 0.2 mL of saline, was administered it. After 14 days, the animals were killed by administration of a nembutal overdose. Lungs were inflated with 4% neutral buffered formalin under a 22 cm formalin pressure. Lungs were embedded in paraffin (5 µm thick sections, stained with hematoxylineosin). Cross sections of the intact lungs were performed on each animal. Signs of pulmonary emphysema were estimated by measure of mean linear intercepts according to Dunnill.31

Acknowledgment. The authors wish to thank Dr. Jean-Paul Vilaine for evaluation of compound 14a on the human LDL peroxidation assay, Dr. Jean-Paul Volland and his group for analytical and spectral studies, and Mrs. Danielle Pommier for expert technical assistance. References (1) For a complete review on HLE and its synthetic inhibitors, see: Edwards, P. D.; Berstein, P. R. Synthetic Inhibitors of Elastase. Med. Res. Rev. 1994, 14, 127-194. (2) Janoff, A. Elastase and Emphysema. Current Assessment of the Protease-Antiprotease Hypothesis. Am. Rev. Respir. Dis. 1985, 132, 417-433. (3) Merritt, T. A.; Cochrane, C. G.; Holcomb, K.; Bohl, B.; Hallman, M.; Strayer, D.; Edwards, D.; Gluck, L. Elastase and R1-PI Proteinase Inhibitor Activity in Tracheal Aspirates during Respiratory Distress Syndrome. J. Clin. Invest. 1983, 72, 656666. (4) Hanley, M. E.; Repine, J. E. Elastase and Oxygen Radicals: synergistic Interactions. Agents Action 1993, Suppl. 42, 39-47. (5) Ishizaki, T.; Kishi, Y.; Sasaki, F.; Takahashi, H.; Ameshima, S.; Onishi, T.; Sakai, T.; Kasamatsu, A.; Nakai, T.; Miyabo, S. Probucol, an oral hypocholesterolemic agent, inhibits acute tobacco smoking induced decrease in plasma antielastase activity and ferroxydase activity in rats. Oxygen Radicals; Elsevier: New York, 1992; pp 573-576. Kim Seow, W.; Thong, Y. H.; Nelson, R. D.; Macfarlane, G. D.; Herzberg, M. C. Nicotine-induced Release of Elastase and Eicosanoids by Human Neutrophils. Inflammation 1994, 18, 119-127. (6) For a short review on ICI 200 880, see: Edwards, P. D. ICI 200 880. Drugs Future 1995, 20, 662-665. (7) Williams, J. C.; Stein, R. L.; Giles, R. E.; Krell, R. D. Biochemistry and Pharmacology of ICI 200,880, a Synthetic Peptide Inhibitor of Human Neutrophil Elastase. Ann. N. Y. Acad. Sci. 1991, 624, 230-243. (8) Hemmi, K.; Shima, I.; Imai, K.; Tanaka, H. Trifluoromethylketone tripeptide derivatives. European Patent EP 0494 071 A2, 1992. (9) Skiles, J. W.; Sorcek, R.; Jacober, S.; Miao, C.; Mui, P. W.; McNeil, D.; Rosenthal, A. S. Elastase Inhibitors containing conformationnally restricted lactams as P3-P2 dipeptide replacements. Bioorg. Med. Chem. Lett. 1993, 3, 773-778. (10) De Nanteuil, G.; Gloanec, P.; Lila, C.; Portevin, B.; Boudon, A.; Rupin, A.; Verbeuren, T. J. New tripeptidic Thrombin Inhibitors. Influence of P2 and P3 Residues on Activity and Selectivity. Bioorg. Med. Chem. 1995, 3, 1019-1024. (11) Portevin, B.; Benoist, A.; Re´mond, G.; Herve´, Y.; Vincent, M.; Lepagnol, J.; De Nanteuil, G. New Prolyl endopeptidase inhibitors: in vitro and in vivo activities of azabicyclo[2.2.2]octane, azabicyclo[2.2.1]heptane, and perhydroindole derivatives. J. Med. Chem. 1996, 39, 2379-2391. (12) Kirschenheuter, G. P.; Cheronis, J. C.; Spruce, L. W. Oxidant sensitive and insensitive aromatic esters as inhibitors of human neutrophil elastase. European Patent, EP 0 405 802 A1, 1991. (13) Poli, S.; Mailland, F.; Coppi, G.; Signorelli, G. Preparation of L-3-[N-[2-(2-thienoyl)thiopropionyl]glycyl]thiazolidine-4-carboxylic acid (P 1517) as an elastase inhibitor and antioxidant. World Patent WO 93 18059, 1992. (14) Vincent, M.; Re´mond, G.; Portevin, B.; Serkiz, B.; Laubie, M. Stereoselective Synthesis of new Perhydroindole Derivative of Chiral Iminoacid, a potent Inhibitor of Angiotensin Converting Enzyme. Tetrahedron Lett. 1982, 23, 1677-1680. (15) Vincent, M.; Pascard, C.; Ce´sario, M.; Re´mond, G.; Bouchet, J. P.; Charton, Y.; Laubie, M. Synthesis and Conformational Studies of Zabicipril (S 9650-3), a potent Inhibitor of Angiotensin Converting Enzyme. Tetrahedron Lett. 1992, 33, 7369-7372.

1918

Journal of Medicinal Chemistry, 1997, Vol. 40, No. 12

(16) Vincent, M.; Re´mond, G.; Laubie, M. Preparation and Therapeutic Use of substituted Azabicycloheptane Carboxylic Acids. French Patent FR 2525604, 1983. (17) Peet, N. P.; Burkhart, J. P.; Angelastro, M. R.; Giroux, E. L.; Mehdi, S.; Bey, P.; Kolb, M.; Neises, B.; Schirlin, D. Synthesis of Peptidyl Fluoromethyl Ketones and Peptidyl R-Keto Esters as Inhibitors of Porcine Pancreatic Elastase, Human Neutrophil Elastase, and Rat and Human Neutrophil Cathepsin G. J. Med. Chem. 1990, 33, 394-407. (18) Epimerization on the C-terminal valine for the two diastereomers of compound 14a was observed either in solution (pH and temperature dependent mechanism) or under solid form (temperature dependent mechanism). (19) Skiles, J. W.; Fuchs, V.; Miao, C.; Sorcek, R.; Grozinger, K. G.; Mauldin, S. C.; Vitous, J.; Mui, P. W.; Jacober, S.; Chow, G.; Matteo, M.; Skoog, M.; Weldon, S. M.; Possanza, G.; Keirns, J.; Letts, G.; Rosenthal, A. S. Inhibition of Human Leucocyte Elastase (HLE) by N-Substituted Peptidyl Trifluoromethyl Ketones. J. Med. Chem. 1992, 35, 641-662. (20) Greenstein, J. P.; Winitz, M. Chemistry of the Amino Acids; Wiley: New York, 1961; Vol. 3, p 2229. (21) Stein, M. M.; Wildonger, R. A.; Trainor, D. A.; Edwards, P. D.; Yee, Y. K.; Lewis, J. J.; Zottola, M. A.; Williams, J. C.; Strimpler, A. M. In vitro and in vivo inhibition of human leukocyte elastase (HLE) by two series of electrophilic carbonyl containing peptides. Proceedings of the American Peptide Symposium, 11th; Escom: Leiden, 1990; pp 369-370. (22) Trolox ) 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid. (23) Skiles, J. W.; Miao, C.; Sorcek, R.; Jacober, S.; Mui, P. W.; Chow, G.; Weldon, S. W.; Possanza, G.; Skoog, M.; Keirns, J.; Letts, G.; Rosenthal, A. S. Inhibition of Human Leucocyte Elastase by N-Substituted Peptides Containing R,R-Difluorostatone Residues at P1. J. Med. Chem. 1992, 35, 4795-4808. (24) For the selectivity studies, compound 14a was evaluated on the following reptors: R1A, R1B, R2A, R2B, R2C, β1, β2, 5HT1A,

Portevin et al.

(25)

(26)

(27)

(28) (29) (30) (31)

5HT1B, 5HT1D, 5HT2C, 5HT3, M1, M2, M3, NK1, NK2, D1, D2, D3, D4, A1, H1, and µ. Finke, P. E.; Shah, S. K.; Ashe, B. M.; Ball, R. G.; Blacklock, T. J.; Bonney, R. J.; Brause, K. A.; Chandler, G. O.; Cotton, M.; Davies, P.; Dellea, P. S.; Dorn, C. P., Jr.; Fletcher, D. S.; O’Grady, L. A.; Hagmann, W. K.; Hand, K. M.; Knight, W. B.; Maycock, A. L.; Mumford, R. A.; Osinga, D. G.; Sohar, P.; Thompson, K. R.; Weston, H.; Doherty, J. B. Inhibition of Human Leucocyte Elastase. 4. Selection of a Substituted Cephalosporin (L-658,758) as a Topical Aerosol. J. Med. Chem. 1992, 35, 3731-3744. Edwards, P. D.; Andisik, D. W.; Strimpler, A. M.; Gomes, B.; Tuthill, P. A. Nonpeptidic Inhibitors of Human Neutrophil Elastase. 7. Design, Synthesis, and in Vitro Activity of a Series of Pyrimidopyrimidine Trifluoromethyl Ketones. J. Med. Chem. 1996, 39, 1112-1124. Stone, P. J.; Lucey, E. C.; Calore, J. D.; Snider, G. L.; Franzblau, C.; Castillo, M. J.; Powers, J. C. The Moderation of Elastaseinduced Emphysema in the Hamster by Intratracheal Pretreatment or Post-treatment with Succinyl Alanyl Alanyl Prolyl Valine Chloromethyl Ketone. Am. Rev. Respir. Dis. 1981, 124, 56-59. Virca, G. D.; Metz, G.; Schnebli, H. P. Similarities between Human and Rat Leucocyte Elastase and Cathepsin G. Eur. J. Biochem. 1984, 144, 1-9. Nuege, J. A.; Aust, S. D. Microsomal lipid peroxidation. In Methods in Enzymology; Fluscter, S., Pucher, L., Eds.; Academic Press: San Diego, CA, 1987; Vol. 52, pp 302-310. Havel, R. J.; Eder, H. A.; Bragdon, J. H. The distribution and chemical composition of ultracentrifugally separated lipoproteines in human serum. J. Clin. Invest. 1955, 34, 1345-1353. Dunnill, M. S. Quantitative methods in the study of pulmonary pathology. Thorax 1962, 17, 320-328.

JM960772Z